Archives of Disease in Childhood, 1983, 58, 697-702

Goodpasture's syndrome: treatment withplasmapheresis, immunosuppression, andanticoagulationM LEVIN, S P A RIGDEN, J R PINCOTT, C M LOCKWOOD, T M BARRATT, ANDM J DILLON

Renal Unit and Department of Histopathology, The Hospital for Sick Children, Great Ormond Street,London, and The Renal Unit, Hammersmith Hospital, Du Cane Road, London

SUMMARY We report 3 children with Goodpasture's syndrome. In 2 children the diagnosis wasdelayed, treatment began late, and they did not recover renal function. In the third child earlydiagnosis and intensive treatment with plasmapheresis, immunosuppression, and anticoagulationresulted in an initial return of renal function after a prolonged period of anuria. In this child,however, although there was no evidence of disease activity, further deterioration of renal functionsubsequently occurred.

Goodpasture's syndrome, a term describing theassociation of glomerulonephritis and lung haemor-rhagel is an uncommon but frequently fatal illness..Although the term originally referred to this clinicalassociation without any implications as to aetiologyor pathogenesis, elucidation of the pathologicalprocesses2-4 has allowed it to be defined moreaccurately by immunological4 5 and pathologicalfeatures.6 The disorder is associated with antibodiesto glomerular basement membrane (oc-GBM) thatcross react with alveolar basement nmembrane.4 7 8The pathogenic role of these antibodies has beenestablished by transfer experiments59 and they maybe identified in the circulation by radioimmunoassayand by indirect immunofluorescence.9-11 The pres-ence in the circulation of o-GBM together with thecharacteristic linear deposition of IgG along theglomerular basement membrane are the immuno-pathological hallmarks of the condition.5 Thesefeatures enable auto-antibody mediated Good-pasture's syndrome to be distinguished as a specificcondition from other pathological processes withboth renal and pulmonary manifestations such asWegener's granulomatosis, polyarteritis, and acutepost-streptococcal glomerulonephritis with pul-monary oedema. Although simultaneous occurrenceof renal and pulmonary manifestations were thediagnostic features of the syndrome as originallydescribed, x-GBM disease may occur with isolatedrenal or pulmonary involvement, and with bothsystems affected at different stages of the illness.512It has been suggested that the term Goodpasture's

syndrome should be replaced by the pathogeneticallydescriptive and broader definition, 'anti-basementmembrane antibody induced glomerulonephritis andpulmonary haemorrhage'. We have, however, withothers,1314 retained the term Goodpasture's syn-drome for reasons of brevity and current familiarity.

Goodpasture's syndrome (or anti GBM glomeru-lonephritis and pneumonitis) usually affects adults.Most series have shown a predominance of youngmen and we are not aware of any reported cases inchildren other than those alluded to by Anandl5 andSiegler.16Although the cause of the ac-GBM antibody pro-

duction is unknown, exposure to hydrocarbon fumesand viral illnesses have been implicated as pre-cipitating events.17 18 A genetic predisposition hasbeen established by the finding of a strong associa-tion of the human leucocyte antigen DRw2 withthe disease,'9 as well as by its occurrence in identicaltwins.20The renal component is frequently characterised

by a severe proliferative glomerulonephritis withextensive crescent formation. Although there isconsiderable variability in the rate of progression ofthe renal disease and some cases with focal glomeru-lar involvement show only very slow deterioration,patients with an appreciable proportion of crescentshave a uniformly bad prognosis. Most largeseries have found that between 44% and 55% ofpatients have died, with a large percentage of thesurvivors requiring long term dialysis and renaltransplantation.5 13

697

copyright. on M

arch 27, 2021 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.58.9.697 on 1 Septem

ber 1983. Dow

nloaded from

698 Levin, Rigden, Pincott, Lockwood, Barratt, and Dillon

After the recognition of the pathogenetic role ofthe oc-GBM antibody attempts were made to treatthe disease with plasmapheresis coupled withimmunosuppression, to remove physically theoc-GBM, and to prevent or diminish their pro-duction. Several published reports have shown thata considerable improvement in mortality may beachieved with control of lung haemorrhage, re-duction in antibody level, and improvement in renalfunction; and plasmapheresis has become standardtreatment for the disorder in adults. Recovery ofrenal function in patients with immunologicallyproved Goodpasture's syndrome is, however, ex-tremely rare once total anuria has occurred.21-23We report our experience with 3 children with

Goodpasture's syndrome. The diagnosis in eachcase was confirmed by renal biopsy examination andthe finding of circulating oc-GBM. In the first 2patients delay in diagnosis and in initiating treatment,together with technical difficulties in treatment,resulted in failure to halt the progress of the disease,and permanent loss of renal function ensued. Inthe third patient early diagnosis and treatment withintensive plasmapheresis, immunosuppression, andanticoagulation resulted in an initial return of renalfunction after a prolonged period of anuria.

Case reports

Case 1. A previously well 4 year old girl was ad-mitted to her local hospital in November 1977 witha 4 month history of loin pain, pallor, anorexia, andlethargy. There was no history of preceding in-fection, hydrocarbon exposure, or a family historyof renal disease. She was anaemic and mildlyuraemic with microscopic haematuria and protein-uria. Renal function deteriorated progressively overthe next few weeks and 4 weeks later she was trans-ferred to the Renal Unit at the Hospital for SickChildren.On admission she was pale and normotensive, but

had no other abnormal respiratory or other signs.Investigation results showed the following: haemo-globin concentration 7.5 g/dl; platelets 160 x 109/l(160 000/mm3); white cell count 12 3 x 109/l(12-3 x 103/mm3) (neutrophils 80%, lymphocytes15 %, monocytes 2%, eosinophils 2 %); fibrin degra-dation products and fibrinogen normal; erythrocytesedimentation rate 70 mm/h; plasma creatinine466 ,umol/l (5*27 mg/100 ml); urea 23 mmol/l(138.5 mg/100 ml); albumin 29 g/l; electrolytesnormal; plasma C3 and C4 normal; antinuclearfactors and anti-deoxyriboneucleic acid antibodiesnot detected; antistreptolysin 0 titre 150 U/ml;Coombs's test negative; no cold agglutinins or im-mune complexes detected; chest radiograph normal;urine showed microscopic haematuria, granular

casts, and heavy proteinuria. A renal biopsy speci-men showed an end stage glomerulonephritis withevidence of previous crescent formation in allglomeruli, and extensive tubular atrophy and inter-stitial fibrosis. Immunofluorescence showed diffuselinear deposition of IgG and smaller amounts of C3in the capillary loops, but staining for IgA, IgM,CIq, and fibrin was negative.A diagnosis of Goodpasture's syndrome was made

on the basis of the typical linear deposition of IgGon immunofluorescence. This was confirmed bydetection of high titres of circulating oc-GBM(>50%; normal range >12% binding of referencepositive serum) using a sensitive solid phase radio-immunoassay.24 She was transferred to Hammer-smith Hospital for plasmapheresis 2 weeks later.There were considerable difficulties in establishingadequate vascular access for plasmapheresis, initiallyvia an arteriovenous shunt and subsequently througha femoral vein line. Attempts at plasmapheresis wereabandoned after 2 exchanges of 500 ml. Treatmentwith prednisolone, azathioprine, and cyclophospha-mide produced no improvement, and she died 6weeks later.

Case 2. A previously well 10 year old girl developedhaematuria and vomiting 1 day after the onset of asore throat and fever for which ampicillin had beenprescribed. There was no other drug or hydrocarbonexposure. She had 1 sister with acute lymphaticleukaemia and another with epilepsy, but no familyhistory of renal disease. Urine output decreased overthe next 2 days and she was admitted to hospitalwhere she was found to be uraemic and oliguric.Renal function deteriorated over the next 3 daysand she was transferred to the Renal Unit at theHospital for Sick Children 6 days after the onset ofsymptoms.On admission she was pale and normotensive, but

there were no other abnormal physical signs. Resultsof investigations were as follows: haemoglobin 11g/dl; white cell count 16-9 x 109/l (16.9 x 103/mm3)(79% neutrophils, 17% lymphocytes and 3%monocytes); coagulation studies normal; plasmacreatinine 400 pumol/l (4.5 mg/100 ml); urea 27mmol/l (163 mg/100 ml); elcctrolytes normal. H(rurine contained numerous white cells, red cells,granular casts, and there was heavy proteinuria.Her chest radiograph was normal. Plasma C3 waslow on 2 occasions (40% and 17% of standard)but C4 and total haemolytic complement werenormal: antinuclear factor, anti-deoxyribonucleicacid antibodies, and immune complexes were notdetected. Throat swab, blood cultures, and viralserology and cultures were all negative, and herantistreptolysin 0 titre was not raised.

copyright. on M

arch 27, 2021 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.58.9.697 on 1 Septem

ber 1983. Dow

nloaded from

Goodpasture's syndrome: treatment with plasmapheresis, immunosuppression, and anticoagulation 699

The low C3 and preceding throat infection sug-gested a diagnosis of acute post-streptococcalglomerulonephritis and an initial period of fluidrestriction and observation was undertaken. Renalfunction progressively deteriorated, however, andrenal biopsy examination was performed 4 daysafter admission. After the biopsy examination shebecame increasingly uraemic and peritoneal dialysisbegan the following day.The renal biopsy histology (Fig. 1) showed an

endocapillary and an extracapillary proliferativeglomerulonephritis with florid epithelial crescents in80% of the glomeruli. There was insufficient materialfor immunofluorescence. Treatment for her crescenticnephritis began with prednisolone 40 mg/M2 per day,azothioprine 2 mg/kg per day, dipyridamole 10mg/kg per day, and heparin 100 U/kg 8 hourly.

Since the cause of the crescentic nephritis wasstill unclear, blood was sent to the HammersmithHospital for oc-GBM titre and the result, receivedwithin 24 hours, showed an extremely high titre(>100% binding) establishing the diagnosis ofGoodpasture's syndrome. Plasmapheresis beganwithout delay (after 8 days of peritoneal dialysis)and was performed daily for the next 4 days,exchanging 1.2-2.0 litres of plasma on eachoccasion. Three days later the oc-GBM titre hadfallen to one quarter of the original value but shedeveloped peritonitis and abdominal wall cellulitisaround the site of the peritoneal dialysis catheter,which had been resited because of drainage problems.

In view of the rapidly spreading infection furtherimmunosuppression and plasmapheresis was feltto be hazardous, and she was transferred forhaemodialysis. No further attempts were made toremove or suppress a-GBM formation with plasma-pheresis or immunosuppression and she has re-rnained in renal failure on long term haemodialysis.

Fig. 1 Renal biopsy, case 2 ( x 450-methenamine silver).The epithelial crescent surrounds the glomerular tuft.

Case 3. A previously well 7 year old girl developeddiarrhoea and vomiting. There was no history ofexposure to hydrocarbons or a preceding infectiveepisode. She had suffered from episodes of mildwheezing and dyspnoea on exercise during thepreceding year.Diarrhoea stopped after 24 hours but she con-

tinued to vomit intermittently. Over the course ofthe next week she became increasingly lethargic,anorexic, and pale and she passed urine less fre-quently. Nine days after the onset of symptoms shewas admitted to hospital. She was found to beanaemic, uraemic, and anuric and was thereforetransferred to the Renal Unit at the Hospital forSick Children on the following day.On admission she was pale and drowsy and her

blood pressure was 110/70 mmHg. She had a moistcough but no abnormal signs in the lungs or cardio-vascular system and there were no other importantfindings. Initial investigation results were as follows:haemoglobin 7.4 g/dl, white cell count 16.6 x 109/l(16-6 x 103/mm3) (84% neutrophils, 11 % lympho-cytes and 3 % monocytes); erythrocyte sedimentationrate 20 mm/h, platelets 500 x 109/l (500 000/mm3);coagulation studies, serum fibrin degradationproducts, and plasma fibrinogen normal; plasmacreatinine 1300 ,umol/l (14.7 mg/100 ml); urea 61mmol/l (367 mg/100 ml); electrolytes normal;albumin 26 g/l; plasma C3 normal; antinuclearfactor, anti-deoxyribonucleic acid antibodies, andimmune complexes not detected; antistreptolysin 0titre 100 units/ml; viral serology negative. Herchest radiograph was normal initially but a repeat 3days later showed a haze over the right lung withperibronchial interstitial shadowing. No urine wasobtainable for examination.

Peritoneal dialysis began soon after admission.Because of a close clinical resemblance to ourprevious patient, blood was sent on the night ofadmission to the Hammersmith Hospital forestimation of o-GBM titre. Twenty four hours laterthe result was received showing an extremely hightitre of oc-GBM (>45% binding), and plasma-pheresis, immunosuppression, and anticoagulationwere begun immediately-without delaying treat-ment to perform a renal biopsy for diagnosticconfirmation.

Plasmapheresis was performed using the Hemo-netic cell separator. The procedure was accom-plished via a percutaneous central line for the first10 days, and thereafter through peripheral veins.She was started on treatment with prednisolone 60mg/M2, cyclophosphamide 3 mg/kg per day, andheparin 100 U/kg 8 hourly. Her progress is illus-trated in Fig. 2.

Plasmapheresis was undertaken daily for 10 days,

copyright. on M

arch 27, 2021 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.58.9.697 on 1 Septem

ber 1983. Dow

nloaded from

700 Levili, Rigden, Pincott, Lockwood, Barratt, and Dillon

Hepirin Warfarin

Cydophosphaide Azath =prinPrednisolonePeritoneol dialysis

O 10 20 30 40 50 60 70 80Days from onset of anuria

Fig. 2 Progress ofpatient in case 3 in relation totreatment.

Normal range for ac-GBM titre <12% binding of reference positiveserum. Urine output not measured after day 42.

exchanging 1 * 3-1.6 1 plasma (about twice thepatient's plasma volume) on each occasion. Theprocedure was performed over 4-6 hours, and waswell tolerated by the patient.

After the first 10 days of treatment, she remainedanuric and the oc-GBM antibody titre on day 10was still notably raised, although lower than pre-treatment value. A more intensive effort was there-fore made to remove the antibody. The volume ofplasma exchanged on each occasion was increasedto between 2 - 8 and 3.8 litres (rai4 times the patient'stotal plasma volume) and a further 11 exchangeswere performed, taking between 6-8 hours on eachoccasion.On day 16 of treatment, after the first 3 exchanges

of 3-8 1, the x-GBM titre fell to within normalvalues and a small quantity of urine was passed.Thereafter there was a progressive increase in urineoutput and the antibody remained low. Dialysis wasdiscontinued on day 24 of treatment and plasma-pheresis stopped after 21 exchanges. A renal biopsyexamination performed on day 27 of treatmentshowed crescentic nephritis affecting all glomeruliwith the characteristic linear deposition of IgG onimmunofluorescence, as shown in Fig. 3.

After the end of plasmapheresis renal functionstabilised with plasma creatinine value of 100-200,umol/l (1-1-2*2 mg/100 ml). Heparinization wasstopped before renal biopsy examination, and afterrecovery from this warfarin was introduced forlong term anticoagulation. oa-GBM antibody re-mained low after stopping plasmapheresis until day48 when it again became notably raised. This rise in

Fig. 3 Renal biopsy, case 3 (x 720). Immunofluorescenceshows a diffuse linear deposition ofIgG in glomerularcapillary loops.

antibody titre was associated with an increase inthe plasma creatinine concentration. Three moreplasma exchanges were performed resulting inundetectable values of a-GBM in the plasmaand a return of plasma creatinine to previousconcentrations.Cyclophosphamide was continued for 8 weeks

and was then withdrawn and replaced by azathio-prine, 2 mg/kg per day. Steroid dosage was de-creased and she was maintained on prednisolone25 mg on alternate days, azathioprine, and warfarin.For the first 6 months after treatment renal

function remained relatively stable, with a glomeru-lar filtration rate of 18 ml/min/1-73 M2. She con-tinued to have proteinuria and mild hypertensionwhich were controlled by prazosin and propranolol,and was able to return to school.

Six months after her initial recovery and despitea-GBM titre remaining undetectable, renal functionbegan to decline. A repeat renal biopsy examinationwas performed in March 1980 to assess whether thisdeterioration was caused by continuing activedisease, or by sclerosis of glomeruli damaged bythe original insult. The biopsy specimen showed noevidence of active cellular proliferation, all glomerulicontained fibrous crescents, and there was extensivetubular atrophy and interstitial fibrosis. Immuno-fluorescence continued to show linear disposition ofIgG. The appearances suggested an end stagecrescentic nephritis with no evidence of active disease.

Despite the negative a-GBM titres and the lack ofhistological evidence of active disease, 6 moreplasmaphereses were performed. There was, how-ever, no improvement in renal function which slowlydeclined, and continuous ambulatory peritonealdialysis began 13 months after her original illness.

copyright. on M

arch 27, 2021 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.58.9.697 on 1 Septem

ber 1983. Dow

nloaded from

Goodpasture's syndrome: treatment with plasmapheresis, immunosuppression, and anticoagulation 701

Thus, despite her dramatic initial improvementafter intensive treatment and the apparently inactivedisease, permanent loss of renal function eventuallyoccurred-presumably caused by sclerosis ofglomeruli damaged by the original insult.

Discussion

These 3 children are among the first recorded casesof immunologically confirmed Goodpasture's syn-drome in childhood, and the third child is also 1 ofthe few patients of any age with this disease to haverecovered renal function once total anuria hadoccurred. Both these unusual events require ex-planation.The lack of previous published reports of

Goodpasture's syndrome in children may have beencaused by failure of recognition rather than in-frequent occurrence of the condition. Childrenthought to have post-streptococcal glomerulone-phritis or crescentic nephritis of unknown aetiologymay have had Goodpasture's syndrome and beenmisdiagnosed because of lack of awareness of itsoccurrence in childhood. Several reports ofGoodpasture's syndrome in adults have commentedon an apparent increase in incidence in recentyears but this probably reflects increased awarenessof the condition and facilities for more accuratediagnosis rather than a changing disease pattern.13The same phenomenon may apply in children butthis explanation does not adequately account for thelack of cases. Since immunofluorescence of renalbiopsy specimens became routine in recent yearsand the linear deposition of immunoglobulin isunlikely to have been overlooked, alternativeexplanations should be considered.Autoimmune disease is rare in childhood, and it

is interesting that Anand et al.V5 reported that theoc-GBM antigen that reacts with Goodpasture's serais absent in young children. The children in cases 2and 3 presented within 1 month of each other. Inview of the known precipitation of Goodpasture'ssyndrome by hydrocarbons and viral infections anunidentified environmental agent may have beenresponsible. Other paediatric centres should bealert to the occurrences of the disease in childhood,as the disease pattern may be changing.The low C3 value in the child in case 2 is interesting

as this is not usually a feature of Goodpasture'ssyndrome. The preceding upper respiratory in-fection, together with the initial low C3 value inthis patient suggested a diagnosis of post-strepto-coccal glomerulonephritis. Streptococcal antigenscross react with GBM antigens,24 and intercurrentinfections enhance any antibody mediated injury.25A preceding streptococcal (or other) infection may

therefore have been responsible for initiating orenhancing the oc-GBM mediated injury and bringingthe case to light.There are several possible reasons for the un-

usual recovery of renal function after 16 daysanuria in the third patient. The kidney in childhoodmay have a greater potential for recovery than inadults but this explanation is inadequate as return ofrenal function was very closely associated withremoval of a-GBM antibody from the circulation.It is therefore likely that recovery was directlyrelated to the effectiveness of treatment in removinghe o-GBM antibody.The management of our third patient differed

considerably from that of our first 2, and of mostreported cases in adults. Lack of awareness of thepossibility of Goodpasture's syndrome in childhoodresulted in both the first 2 patients undergoing aperiod of observation for presumed acute glomeru-lonephritis, and renal biopsy examination was onlyperformed when continued deterioration of renalfunction occurred. The delay before referral,coupled with the period of observation and timespent awaiting the result of the renal biopsy ex-amination resulted in a total delay of 2 months inthe first patient and 2 weeks in the second, beforetreatment began. In addition to the delay beforetreatment lack of familiarity with the technicalaspects of plasmapheresis in small ill childrenresulted in less effective treatment in these patients.

In contrast the diagnosis in the third patient wasestablished and treatment with plasmapheresis,immunosuppression, and anticoagulation beganwithin 24 hours of hospital admission. The avail-ability of a rapid and sensitive -.GBM assay madequick diagnosis and treatment possible withoutfirst performing a renal biopsy examination. Thespeed with which measures to remove the o-GBMbegan may have been a crucial factor in this patient'srecovery.The technical difficulties in performing plasma-

pheresis on small ill children had largely been over-come by the time our third patient was admitted.This enabled us to perform plasmapheresis rapidlyand more intensively. The initial exchange ofvolumes equivalent on body weight basis to thoseused in most reported adult cases produced only aslow decrease in ac-GBM titre. Increasing thevolume exchanged daily to 4 times the patient'stotal plasma volume resulted in a rapid drop ofax-GBM and a concurrent return of renal function.In addition to removing antibody, plasmapheresismay be beneficial in removing mediators of in-flammation such as complement, fibrinogen, andkinins. The greater volumes exchanged would haveproduced more effective removal of these mediators

copyright. on M

arch 27, 2021 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.58.9.697 on 1 Septem

ber 1983. Dow

nloaded from

702 Levin, Rigden, Pincott, Lockwood, Barratt, and Dillon

and may also have played some role in the improvedoutcome.

If these explanations are correct and the recoveryof renal function in the third child was due to moreeffective and rapid removal of the pathogenicantibody, a similar approach should be used inother patients. Although our patient eventuallydeveloped irreversible renal failure, her initial un-expected recovery suggests that earlier and moreintensive treatment may produce complete recoveryin other patients, particularly in those treated beforeextensive damage has occurred.

Because of the need to begin treatment quickly ifpermanent loss of renal function is to be averted,the diagnosis of Goodpasture's syndrome needs tobe considered early in patients with nephritis withor without pulmonary symptoms. There are con-siderable logistic problems in measuring a-GBMtitres in all patients with glomerulonephritis but asthe incidence of post-streptococcal nephritis hasdeclined in most developed countries the rarer andmore ominous forms of glomerulonephritis con-stitute a greater proportion of cases. Any child withan acute nephritic syndrome who has either severerenal dysfunction or who fails to follow the expectedpathway to recovery should be referred urgently toa children's renal unit and be investigated immuno-logically and by renal biopsy examination as thereare new methods of treatment that may avertpermanent loss of renal function.

ML was supported by grants from the Kidney Research AidFund and the National Kidney Research Fund. SPAR wassupported by the Kidney Research Aid Fund.

References

Stanton MC, Tange JD. Goodpasture's syndrome (pul-monary haemorrhage associated with glomerulone-phritis). Austraiasian Annals ofMedicine 1958 ;7:132-44.

2 Scheer RL, Grossman MA. Immune aspects of theg!omerulonephritis associated with pulmonary hemor-rhage. Ann Intern Med 1964;60:1009-21.

3Lerner RA, Glassock RJ, Dixon FJ. The role of anti-glomerular basement membrane antibody in the patho-genesis of human glomerulonephritis. J Exp Med1967;126:989-1004.

4Markowitz AS, Battifora HA, Schwartz F, Aseron C.Immunological aspects of Goodpasture's syndrome.Clin Exp Immunol 1968 ;3:585-91.

5Wilson CB, Dixon FJ. Anti-glomerular basement mem-brane antibody induced glomerulonephritis. Kidney Int1973 ;3 :74-89.

6 Dunnhill MS. Goodpasture's syndrome. In: Dunnill MS,ed. Pathological basis of renal disease. London: WBSaunders, 1976:49-57.

7Heptirnstall RH. Lung hemorrhage and glomerulone-phritis or Goodpasture's syndrome. In: Heptinstall RH,

eds. Pathology of the kidney. 2nd ed. Boston: Little,Brown & Co, 1974:580-99.

8 Koffler D, Sandson J, Carr R, Kunkel HG. Immunologicstudies concerning the pulmonary lesions in Good-pasture's syndrome. Am JPathol 1969 ;54:293-306.

9Wilson CB, Dixon FJ. The renal response to immuno-logic injury. In: Brenner BM, Rector FC, eds. The kidney.Philadelphia: WB Saunders, 1981 :1237-350.

10 Wilson CB, Marquardt H, Dixon FH. Radioimmuno-assay (RIA) for circulating antiglomerular basementmembrane (GBM antibodies) (Abstract). Kidney Int1974;6:114A.

1 McPhaul JJ, Jr, Dixon FJ. The presence ofanti-glomerularbasement membrane antibodies in peripheral blood.JImmunol 1969 ;103:1168-75.

12 Sissons JGP, Evans DJ, Peters DK, et al. Glomerulone-phritis associated with antibody to glomerular basementmembrane. Br MedJ 1974 ;4:11-4.

13 Teague CA, Doak PB, Simpson IJ, Rainer SP, Herd-son PB. Goodpasture's syndrome: an analysis of29 cases.Kidney Int 1978 ;13:492-504.

14 Whitworth JA, Lawrence JR, Meadows R. Goodpasture'ssyndrome: a review of nine cases and an evaluation oftherapy. Aust NZJ Med 1974 ;4:167-77.

15 Anand SK, Landing BH, Heuser ET, Olson DL,Grushkin CM, Lieberman E. Changes in glomerularbasement membrane antigen(s) with age. J Pediatr 1978;92:952-3.

16 Siegler RL, Bond RE, Morris AH. Treatment ofGoodpasture's syndrome with plasma exchange andimmunosuppression. Clin Pediatr (Phila) 1980;19:488-91.

17 Benoit FL, Rulon DB, Theil GB, Doolan PD, Watten RH.Goodpasture's syndrome, a clinicopathologic entity. AmJMed 1964 ;37:424-44.

18 Beirne GJ. Goodpasture's syndrome and exposure tosolvents. (Editorial). JAMA 1972;222:1555.

19 Rees AJ, Peters DK, Compston DAS, Batchelor JR.Strong association between HLA-DRW2 and antibody-mediated Goodpasture's syndrome. Lancet 1978 ;i :966-8.

20 D'Apice AJF, Kincaid-Smith P, Becker GJ, Lough-head MG, Freeman JW, Sands JM. Goodpasture'ssyndrome in identical twins. Ann Intern Med 1978;88:61-2.

21 Lockwood CM, Person TA, Rees Al, Evans DJ,Peters DK, Wilson CB. Immunosuppression and plasma-exchange in the treatment of Goodpasture's syndrome.Lancet 1976;i:71 1-5.

22 Rosenblatt SG, Knight W, Bannayan GA, Wilson CB,Stein JH. Treatment of Goodpasture's syndrome withplasmapheresis. AmJ Med 1979 ;66 :689-96.

23 Johnson JP, Whitman W, Briggs WA, Wilson CB.Plasmapheresis and immunosuppressive agents in anti-basement membrane antibody-induced Goodpasture'ssyndrome. Am J Med 1978 ;64:54-9.

24 Markowitz AS, Lange CF, Jr. Streptococcal relatedglomerulonephritis. 1. Isolation, immunochemistry andcomparative chemistry of soluble fractions from type 12nephritogenic streptococci and human glomeruli. JImmunol 1964 ;92 :565-75.

25 Rees AJ, Lockwood CM, Peters DK. Enhancedallergic tissue injury in Goodpasture's syndrome byintercurrent bacterial infection. Br MedJ 1977;ii:723-26.

Correspondence to Dr Michael Levin, Renal Unit, Instituteof Child Health, 30 Guildford Street, London WC1N IEH.

Received 27 April 1983

copyright. on M

arch 27, 2021 by guest. Protected by

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.58.9.697 on 1 Septem

ber 1983. Dow

nloaded from